Calculation method of animal transfer behavior and application thereof

Abstract

The invention relates to the field of video acquisition and image analysis and identification, in particular to a method for identifying and calculating a transfer behavior of a tested animal under the driving of a specific disease and a drug by using machine vision. The calculation method of the animal transfer behavior comprises the following steps of recording an image; selecting one frame from the image, and counting the gray level distribution; calculating a threshold value capable of distinguishing an animal from a background, and binarizing each frame of image by taking the threshold value as a boundary; calculating a normalized second-order central moment, obtaining the angle of the animal in each frame, and obtaining an angle array; taking a first-order derivative of the angle array, respectively counting the total number of the positive peaks and the negative peaks, and obtaining a half of the difference value between the total number of the positive peaks and the total number of the negative peaks as the net transfer number of the animal through a method of not counting digits after decimal points. The method is small in error and accurate in data, is completely free of intervention to the animals and is capable of replacing the manual counting.

Description

technical field [0001] The invention relates to the fields of video acquisition and image analysis and recognition, in particular to a recognition and calculation method for the circle behavior of tested animals driven by specific diseases and drugs by using machine vision. Background technique [0002] There are about 600 million patients with nervous system diseases such as Parkinson's disease, cerebral ischemia, brain tumor and spinal cord injury in the world, and with the acceleration of population aging and the influence of multiple factors such as economy, society and environment, The number of patients with these diseases is increasing year by year, which has caused a huge burden on the development of society and economy. [0003] For years, neuroscientists and neurologists have sought new research tools and treatments for the nervous system, including Parkinson's disease, cerebral ischemia, brain tumors and spinal cord injuries. In order to study the relationship be...

AI Technical Summary

Problems solved by technology

[0004] Early monitoring of animal rotation behavior was mainly based on direct observation and manual recording. These methods have many problems such as qualitative subjectivity, time-consuming and labor-intensive.

Most of the existing automatic counting devices need to paste or bind magnets or paste color marks on the surface of animals, which has a certain impact on the natural behavior of animals and also brings certain difficulties to experimental operations.

At the same time, the most common magnet-type counter needs to be equipped with a specific magnetic rotor sensing device, so there are thresholds in equipment cost and hardware

Devices that adopt machine vision recognition and video camera tracking have lower hardware requirements, but the algorithms used in existing commercial products are based on the calculation of the center of mass motion trajectory, which cannot accurately identify specific drugs and diseases driven by more intense diseases. Small wheelbase circling behavior, for example, when unilateral Parkinson's disease model mice have strong lesions, they often show the behavior of circling in situ with a wheelbase smaller than the body axis length or even zero after being stimulated by drugs

Images